Violet Grove CO2 monitoring project Overview of the Violet Grove CO2 seismic monitoring project Don Lawton, Marcia Coueslan, Fuju Chen, Henry Bland, Michael Jones1, Eric Gallant and Malcolm Bertram ABSTRACT In March 2006, the baseline seismic program was completed at the Penn West CO2 injection and monitoring pilot in west-central Alberta. The reservoir is the Cardium Formation in the Pembina field, and the reservoir depth at the pilot site is 1640 m. High- fold multicomponent seismic data were collected along three 2D lines (two parallel, one orthogonal) and from eight triaxial geophones that were cemented into an observation well at the site. All geophones were kept live for all shots of the program, and data quality was excellent. The surface seismic data were processed as individual 2D lines as well as a sparse 3D survey; the data from the downhole geophones were processed as walkaway vertical seismic profiles. The reservoir is a low-amplitude event on P-P and P- S sections and data volumes and shows little structure. Few passive seismic events were recorded in the first 6 months of monitoring. The first monitor survey is scheduled for January, 2006. INTRODUCTION The geological storage of CO2 is a technically feasible way of making significant reductions in emissions of CO2 into the atmosphere. The Intergovernmental Panel on Climate Change (IPCC) first reported on geological storage in its Third Assessment Report (IPCC, 2001), and is currently undertaking a Special Report on CO2 capture and storage, due for release in 2005. The International Energy Agency (IEA) has also completed several studies that discuss opportunities for CO2 sequestration (e.g. IEA GHG, 2001). The WCSB has enormous potential for the geological storage of CO2. Bachu and Shaw (2004) report that ~4 Gt of CO2 could be stored in depleted oil and gas reservoirs in western Canada. In deep saline aquifers, the potential storage capacity is much greater, estimated to 4,000 Gt (Bachu and Adams, 2003). The total storage capacity is very large compared with Canada’s annual CO2 emissions (700 Mt in 2000). However, security of storage is a key issue, and monitoring and verification protocols are required to enable geological storage of CO2 for accounting purposes related to emissions trading scenarios (Chalaturnyk and Gunter, 2004), as well as for acceptance by the general public. Theoretical and laboratory studies of changes in rock properties with CO2 flooding predict that changes in seismic attributes should be observable in field seismic surveys (e.g., Wang et al., 1998; Sinartio, 2002). Time-lapse seismic surveys have been shown to be effective at mapping the CO2 injection plume at the Sleipner CO2 storage site in Norway (Skov et al., 2002; Arts et al., 2002) and at the Weyburn CO2 injection project in southeastern Saskatchewan (White et al., 2004; Li, 2003; Davis et al., 2003). At both of these sites, P-wave amplitude and traveltime anomalies have identified the distribution of CO2 in the reservoir. 1 Schlumberger Canada Ltd. CREWES Research Report — Volume 17 (2005) 1 Lawton et al. In order to properly map the movement of the CO2 plume in the injection reservoir and to track possible leakage paths, three-dimensional (3D) seismic surveys are required. However, 3D surveys with close line spacing and small shot and receiver intervals are expensive, and surface seismic data may have insufficient bandwidth to resolve thin (< 20 m) injection zones. At the Penn West CO2 – enhanced oil recovery project in Alberta, an innovative seismic monitoring strategy has been implemented involving a sparse, multicomponent surface seismic program integrated with active and passive monitoring using geophones permanently cemented into an observation well. The surface seismic program provides 3D subsurface coverage of the pilot site whilst data from the downhole geophones provide high-resolution images around the observation well. The downhole geophone installation will be used for passive monitoring of CO2 injection between active-source seismic surveys. The Penn West baseline survey was completed in March 2005 and passive monitoring has been undertaken for 5 months. The first monitor survey is scheduled for January, 2006. LOCATION AND PROGRAM The Penn West pilot is located in the Pembina Oil Field in west-central Alberta (Twp 48, R8-9 W5M). The primary purpose of the CO2 injection program is for enhanced oil recovery from the field, with a secondary objective of evaluating this depleted reservoir for the geological storage of CO2. The reservoir in this field is the Upper Cretaceous Cardium Formation which occurs at a depth of approximately 1650 m below ground surface. The Cardium Formation is made up of sandstone sheets and a thin conglomerate layer sandwiched between thick black marine shales of the Blackstone Formation (below) and the Wapiabi Formation (above). The total thickness of the Cardium Formation at the site is approximately 20 m. Surface access at the site is limited due to swamps and tree cover. The surface seismic program consisted of two parallel, multicomponent 2D lines, 400 m apart and oriented east-west, and one orthogonal multicomponent 2D north-south line, intersecting near the CO2 injector wells. These three lines were 3 km long in order to yield a good distribution of source-receiver offsets at the injection pad. Two additional short north-south receiver lines were also included to provide added seismic coverage around the injection zone. A detailed map showing wells, seismic lines, and surface culture is shown in Figure 1. SURFACE SEISMIC PROGRAM Data acquisition The seismic program was coordinated by OutSource Seismic Consultants Ltd and data acquisition by undertaken by Veritas Geoservices Ltd on March 15, 2005. Some limited line clearing was required for setbacks from pipelines and infrastructure, and a helicopter was used for distribution of the seismic acquisition equipment along the lines. Acquisition parameters for the survey are summarised in Table 1. 2 CREWES Research Report — Volume 17 (2005) Violet Grove CO2 monitoring project FIG. 1. Map showing Penn West CO2 injection site. Multicomponent seismic lines (Lines 1, 2, 3) are shown in yellow , and receivers-only lines are shown in red R2, R2). A regional seismic line (Line 4) is shown is green. The observation (VSP) well is shown by the filled green circle, and the CO2 injection pad is indicated by the red square. Table 1. Acquisition parameters for the Penn West CO2 seismic monitoring program Acquisition parameter Value Source spacing 40 m Source type Dynamite Source depth 15 to 18 m Receiver spacing 20 m Receiver type Sercel DSU 3C Instruments Sercel 408 XL Sample interval 1 ms A total of 12.8 km of lines were occupied by receivers, and 9.0 km with shots. The total program consisted of 323 source points and 643 receiver points. All lines were kept live for all shots. Figure 2 shows an example of seismic equipment laid out along Line 1. CREWES Research Report — Volume 17 (2005) 3 Lawton et al. FIG. 2. Seismic recording equipment along Line 1. Figures 3 and 4 show a Sercel 3C DSU geophone unit before and after deployment. Considerable care is required in laying out this equipment. The unit is planted in the ground on a spike in an augured hole and is oriented towards magnetic north and is also leveled. FIG. 3. Veritas Sercel DSU 3C unit prior to deployment (with CREWES rep. Dr. Zoulin Chen) 4 CREWES Research Report — Volume 17 (2005) Violet Grove CO2 monitoring project FIG. 4. Sercel DSU 3C unit after deployment In order to minimize noise from wind and the nearby highway 620, data recording was undertaken during the evening. Conditions were ideal with calm winds and occasional light snow. Recording commenced at 7 pm in March 19 and was completed at approximately 11 pm the same evening. University of Calgary representatives (Lawton, Bertram, Gallant, Chen and Coueslan) were on site during data acquisition. Data quality is excellent. Data processing The surface seismic data were processed by Veritas Geoservices Ltd during the spring of 2005. Lines 1 through 3 were initially processed as individual 2D lines. In addition to these lines, recorded for this program, Penn West donated the processing of Line 4, which was an older, 7 km long regional line that cuts diagonally across the northern side of the injection site (location shown by the green line in Figure 1). This line will be used in the development of a regional geological model of the injection site in Phase 2 of this monitoring project. The processing flow used for data processing is listed in Table 2. CREWES Research Report — Volume 17 (2005) 5 Lawton et al. Table 2. Data processing flow and parameters Processing parameters Reformat Geometry assignment Tilt correction to vertical Ground roll attenuation – polarization filter Trace edits (manual) Amplitude recovery (exponent = 2.2) Minimum phase deconvolution (60 ms operator length; 0.1% prewhitening) Tomographic structure statics (datum 910 m; replacement velocity 2500 m/s) Trace gather Velocity analysis Surface consistent residuals statics Spectral whitening (5 – 160 Hz) Mute and final trim statics Surface consistent scaling CDP stack F-X noise attenuation Poststack Kirchhoff migration (100% stacking velocities) Bandpass filter (5-10-100-120 Hz) Scaling (mean, 500 ms window) The final, processed sections for Lines 1 through 4 are displayed in Figures 5 through 8 respectively. The intersection of Line 1 ties very well with Lines 2 and 3. 6 CREWES Research Report — Volume 17 (2005) Violet Grove CO2 monitoring project FIG. 5. Final P-P section, Line 1. The Cardium Fm is identified by the arrow CREWES Research Report — Volume 17 (2005) 7 Lawton et al. FIG. 6. Final P-P section, Line 2. The Cardium Fm is identified by the arrow 8 CREWES Research Report — Volume 17 (2005) Violet Grove CO2 monitoring project FIG.